Performance of Large Language Models in Patient Complaint Resolution: Web-Based Cross-Sectional Survey.

This study provides significant evidence supporting the effectiveness of LLMs in resolution of patient complaints. ChatGPT demonstrated superiority in terms of response appropriateness, empathy, quality, and overall satisfaction when compared against actual human responses to patient complaints. Future research can be done to measure the degree of improvement that artificial intelligence generated responses can bring in terms of time savings, cost-effectiveness, patient satisfaction, and stress reduction for the health care system.